David A. Johnson, MD


March 16, 2012

This feature requires the newest version of Flash. You can download it here.

Colon Cancer and Lynch Syndrome

Hello, I am Dr. David Johnson, Professor of Medicine and Chief of Gastroenterology at Eastern Virginia Medical School in Norfolk, Virginia.

It is March, traditionally known as "colon cancer prevention month." Colon cancer is very prevalent in the United States, if not worldwide. More than 146,000 cases were reported last year alone in the United States. As we strive to make improvements on adherence to screening guidelines and getting our patients in for colonoscopy and appropriate testing, one thing that perhaps we are not doing well enough is recognizing syndromic cancers, such as Lynch syndrome.

Considering Lynch Syndrome

I would like to heighten your awareness and discuss the key elements of suspicion for when you should be considering this condition in your patients. Reflect on your population. If you are a healthcare provider, have you had any patients with colon cancer, ovarian cancer, uterine cancer, small bowel cancer, cancer of the ureter or the renal pelvis, or gastric cancer? Have you thought about Lynch syndrome in these patients?

How many patients do you follow in your clinic populations who have Lynch syndrome? I would bet that it is fairly rare. Retrospective data[1] show that Lynch syndrome is frequently underdiagnosed, even when patients with colorectal cancer meet the criteria for this syndromic cancer. People just aren't thinking about syndromic cancer when one of these cancers is identified.

It is healthy, during colon cancer prevention month, to expand our diagnosis -- not just for screening, but also to appropriate recognition and management of syndromic cancer.

Once thought to be fairly rare, it's estimated that 2%-3% of colon cancers in the United States, and about 3% of uterine cancers, are related to Lynch syndrome. Do the math: 2%-3% of 146,000 -- that's a lot of cancers. Uterine cancer is similar. It is, unfortunately, not an uncommon disease in the United States, so it potentially affects a lot of patients and may not be recognized.

Genetics of Lynch Syndrome

Why is this important? Lynch syndrome is an autosomal-dominant syndromic cancer caused by a genetic defect in one of the 6 mismatch repair genes. The preponderance of the mismatch repair genetic abnormalities for Lynch syndrome fall into 3 types, 2 of which command 90% of the syndromic Lynch patients. This is important because these genetic defects create havoc in the "spell-checker" genes, whose job it is to cut mismatched DNA sections during the normal process of DNA sequencing.[2]

This occurs not infrequently in the genetic replication cycle of the cell. Replication mismatches can occur at a rate of 1 in 106 base pairs per cell division. Mismatch repair ("spell-checker") genes clip these defects out so they don't lead to more aberrant proliferative behavior such as cancer.

Clinical Implications

Lynch syndrome is a genetic defect that has significant implications on the clinical side. A patient whose genetic testing is positive for Lynch syndrome has a likelihood of developing colon cancer that approaches 70% over his or her lifetime. The risk for uterine cancer is similar or a bit higher, and the risk for ovarian cancer is13%-15%. Almost certainly, these patients will have some type of cancer during their lifetime.

Metachronous cancers occur in 7%-10% of patients with Lynch syndrome. These patients present with one cancer, and typically in the course of a short period of time, another cancer develops. In patients with Lynch syndrome who have colon cancer, if you don't perform a subtotal colectomy because you didn't recognize that it was Lynch syndrome, the likelihood that the patient will develop a repeat colon cancer over 30 years (even in patients who are being screened) approaches 65%. This is not uncommon. A lot of these diagnoses are missed, and a secondary cancer develops.

Criteria for Testing for Lynch Syndrome

When should you start thinking about Lynch syndrome? Criteria were established in the early 1990s called the Amsterdam criteria, with the famous "3-2-1 rule." Patients had 3 relatives with Lynch-type cancers, 2 of which were in different generations, and 1 of which was in a patient who was younger than 50 years. Unfortunately, this was not very sensitive nor very specific (sensitivity less than 40% and specificity less than 70%), so these criteria didn't work very well.

A refinement of these criteria is known as the Bethesda criteria, which had an iteration in the mid-1990s and a second iteration in the early 2000s with a more specific definition of patients with microsatellite instability.[3] The Bethesda criteria describe a patient who had colon cancer under the age of 50 years; or who had colon cancer themselves plus a family member with a Lynch-related cancer (uterine cancer, bladder cancer, transitional cell cancer of the ureter, ovarian cancer, gastric cancer, and small bowel cancer). These are cancers that certainly wouldn't be thought of as traditional links with colon cancer unless you were considering Lynch syndrome as a potential diagnosis.

The Bethesda criteria also looked at histologic criteria, on which you can consult with your pathologist. A Crohn's type of picture, in which patients have lymphocytes infiltrating into the tumor, is a hallmark, or is at least suspicious, for Lynch syndrome. Medullary pattern or mucinous or signet differentiation patterns are also hallmarks, and the pathologist should start thinking about variants of Lynch syndrome.

Testing for Lynch Syndrome

How do you test for Lynch syndrome at the pathology level? One thing you should request is an evaluation by immunohistochemical staining for microsatellite instability. Basically, this is a test that looks for the loss of the alleles of the mismatch repair genes. They can test for these by special stain. If the stain is suggestive, then you can do the specific genetic blood test looking for the markers that are consistent with Lynch syndrome.

Once the patient tests positive for an aberration of the mismatch repair gene, it is very easy to profile that patient and develop a pedigree to include each first-degree relative who might benefit from testing and education about Lynch syndrome.

You need to start thinking about syndromic cancers because the relative risk is quite high. You need to be thinking about this in patients with early cancers -- not just colon cancer, but in uterine cancer and ovarian cancer as well.

The onset of colon cancer associated with Lynch syndrome seems to be shifted forward at least a decade, if not more. In the United States, the average age for sporadic cancer of the colon is approximately 65 years. If you look at colon cancer in Lynch syndrome, the average age of onset is earlier (45 years) and some patients may be in their 20s. For sporadic uterine cancer, the average age is approximately 60 years, but in Lynch syndrome, it is shifted by a decade, to 50 years. When you see these cancers in a younger patient, the bell should go off and you should start thinking about Lynch syndrome and doing the appropriate testing.

Cost-Effectiveness of Testing

Work closely with your pathologist. Ask the pathologist to test these patients when they have resections for colon cancer. A study published last year in the Annals of Internal Medicine[4] suggested that it was very cost-effective to check all colon cancers for microsatellite instability. If the test is negative, do a BRAF oncogene check to improve identification of Lynch syndrome. The cost-effectiveness was about $36,000 per life-year saved, very comparable with other tests that are viewed as being cost-effective.

It is important to think about extracolonic cancers in these patients and to start to put the dots together because we are missing the boat on a lot of these syndromic cancers. It's not a zebra; it's 2%-3% of the patients who are presenting with cancer. We need to think outside the box. Even if you are not a gastroenterologist, include the colon when you see syndromic related cancers of the uterus, ovary, small bowel, or stomach.

These things start to come together, and you can only make the diagnosis when you start thinking about it. As we approach colon cancer screening month, hopefully we will do a better job when we start thinking about syndromic cancers.

I look forward to talking to you again soon. Thanks for listening.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: